LightCMR-Bench

LightCMR-Bench is a feature-level cross-modal retrieval benchmark and reviewer artifact package for TGRC-PQ. It provides pre-extracted image and text embeddings derived from the Localized Narratives benchmark, together with supplementary files used to reproduce the TGRC-PQ experiments.

This repository is not a load_dataset()-style tabular dataset. The Hugging Face Dataset Viewer is intentionally disabled because the repository contains NumPy feature matrices, Faiss indexes, PyTorch checkpoints, JSON/CSV result files, and qualitative evidence files.

What Is Included

embeddings_large_d1024/
  {coco_retrieval,flickr30k,open}/
    {train,val,test}/
      image.npy        # 1024D image features
      text.npy         # 1024D text features
      ids.json         # sample identifiers

embeddings_mini_d64/
  {coco_retrieval,flickr30k,open}/
    {train,val,test}/
      image.npy        # 64D image features
      text.npy         # 64D text features
      ids.json         # sample identifiers

openreview_artifacts/
  README.md
  faiss_models/                         # saved PQ/OPQ/RQ Faiss indexes and metadata
  checkpoints/locked_main_table/         # TGRC-PQ checkpoints and per-run logs
  supplementary_experiments_raw/         # raw ablation and sensitivity outputs
  qualitative_rank_evidence/             # ranking evidence and selected examples
  results/                               # compact summaries used by the paper

The previous root-level coco_retrieval/, flickr30k/, and open/ folders have been removed. Please use the namespaced embedding folders above.

Data Provenance

The files in this repository are derived feature files, not the original images or captions. The underlying image-text pairs come from the Localized Narratives benchmark:

https://google.github.io/localized-narratives/

We extracted 64D student features and 1024D teacher features with BEiT3-Large from the Flickr30K, COCO, and OpenImages portions of Localized Narratives, and use these feature files to build LightCMR-Bench for cross-modal retrieval and quantized-retrieval experiments.

Please cite the original Localized Narratives paper when using these artifacts. This repository should not be treated as a replacement for the original dataset or its license terms.

Quick Download

Download only the 64D features:

from huggingface_hub import snapshot_download

snapshot_download(
    repo_id="zhoukun/LightCMR-Bench",
    repo_type="dataset",
    allow_patterns="embeddings_mini_d64/**",
    local_dir="LightCMR-Bench",
)

Download only the 1024D features:

from huggingface_hub import snapshot_download

snapshot_download(
    repo_id="zhoukun/LightCMR-Bench",
    repo_type="dataset",
    allow_patterns="embeddings_large_d1024/**",
    local_dir="LightCMR-Bench",
)

Download the TGRC-PQ reviewer artifacts:

from huggingface_hub import snapshot_download

snapshot_download(
    repo_id="zhoukun/LightCMR-Bench",
    repo_type="dataset",
    allow_patterns="openreview_artifacts/**",
    local_dir="LightCMR-Bench",
)

Using the Artifact Bundle with the Code Repository

In the TGRC-PQ code repository, create symlinks to the downloaded artifact folders:

mkdir -p data
ln -s /path/to/LightCMR-Bench data/lightcmr_bench
ln -s /path/to/LightCMR-Bench/openreview_artifacts/faiss_models faiss_models

For checkpoint inspection:

mkdir -p runs/locked_main_table
ln -s /path/to/LightCMR-Bench/openreview_artifacts/checkpoints/locked_main_table runs/locked_main_table/calibrated

The feature cache under runs/faiss_feature_cache is intentionally not included. It is deterministic cache data produced by the training script and can be regenerated from the feature files and Faiss models.

File Sizes

Approximate remote sizes:

embeddings_large_d1024: 6.39 GiB
embeddings_mini_d64:   0.58 GiB
openreview_artifacts:  0.16 GiB

Citation

TGRC-PQ paper:

Zhou, Kun and HASSAN, FADRATUL HAFINAZ and Gan, Keng Hoon,
Attention-Guided Product Quantization for Efficient Cross-Modal Retrieval.
Available at SSRN: https://ssrn.com/abstract=6022619 or
http://dx.doi.org/10.2139/ssrn.6022619

@misc{zhou_attention_guided_pq_ssrn,
  title        = {Attention-Guided Product Quantization for Efficient Cross-Modal Retrieval},
  author       = {Zhou, Kun and Hassan, Fadratul Hafinaz and Gan, Keng Hoon},
  howpublished = {Available at SSRN: \url{https://ssrn.com/abstract=6022619}},
  doi          = {10.2139/ssrn.6022619}
}

Localized Narratives:

@inproceedings{ponttuset2020connecting,
  title     = {Connecting Vision and Language with Localized Narratives},
  author    = {Pont-Tuset, Jordi and Uijlings, Jasper and Changpinyo, Soravit and Soricut, Radu and Ferrari, Vittorio},
  booktitle = {European Conference on Computer Vision},
  pages     = {647--664},
  year      = {2020},
  organization = {Springer}
}

LightCMR-Bench / TGRC-PQ artifacts:

@misc{zhoukun_lightcmr_bench,
  title        = {LightCMR-Bench: Feature-Level Cross-Modal Retrieval Benchmark and TGRC-PQ Artifacts},
  author       = {Zhoukun},
  howpublished = {\url{https://huggingface.co/datasets/zhoukun/LightCMR-Bench}},
  note         = {Derived features from Localized Narratives extracted with BEiT3-Large}
}